1. NUCLEAR SYSTEMS ENHANCED PERFORMANCE PROGRAM
Technical Lessons
Learned from the
Fukushima-Daichii
(617) 452-2660
canes@mit.edu
mit.edu/canes
(617) 452-2660
canes@mit.edu
mit.edu/canes
Accident and Possible
Corrective Actions for the
Nuclear Industry:
An Initial Evaluation
N
N
Massachusetts Institute of Technology
Massachusetts Institute of Technology
77 Massachusetts Avenue, 24-215
77 Massachusetts Avenue, 24-215
S
S
Cambridge, MA 02139-4307
Cambridge, MA 02139-4307
J. Buongiorno, R. Ballinger, M. Driscoll,
B. Forget, C. Forsberg, M. Golay,
M. Kazimi, N. Todreas, J. Yanch
A
A
E
E
C
C
CANES
CANES
MIT-NSP-TR-025
May 2011

2.
2

3.
TABLE
OF
CONTENTS
PREMISE
5
OBJECTIVES
OF
THE
REPORT
5
1.
EMERGENCY
POWER
FOLLOWING
BEYOND-­‐DESIGN-­‐BASIS
EXTERNAL
EVENTS
7
2.
EMERGENCY
RESPONSE
TO
BEYOND-­‐DESIGN-­‐BASIS
EXTERNAL
EVENTS
7
3.
HYDROGEN
MANAGEMENT
9
4.
CONTAINMENT
10
5.
SPENT
FUEL
POOLS
10
6.
PLANT
SITING
AND
SITE
LAYOUT
11
A
FEW
CLOSING
THOUGHTS
13
APPENDIX
A
–
PUBLIC
HEALTH
IMPACT
OF
FUKUSHIMA
14
APPENDIX
B
-­‐
RESPONSE
OF
THE
U.S.
NRC
AND
NUCLEAR
INDUSTRY
TO
THE
FUKUSHIMA
ACCIDENT
18
3

4.
4

5. CANES PUBLICATIONS
Topical and progress reports are published under six series:
Advanced Nuclear Power Technology Program (MIT-ANP-)
Nuclear Fuel Cycle Technology and Policy Program (MIT-NFC-)
Nuclear Systems Enhanced Performance Program (MIT-NSP-)
Nuclear Energy and Sustainability Program (MIT-NES-)
Nuclear Space Applications (MIT-NSA-)
MIT Reactor Redesign Program (MIT-MRR)
Please visit our website (mit.edu/canes/) to view more publication lists.
MIT-NSP- Series :
MIT-NSP-TR-025 J. Buongiorno, et al., Technical Lessons Learned from the Fukushima-Daichii
Accident and Possible Corrective Actions for the Nuclear Industry :
An Initial Evaluation (May 2011).
MIT-NSP-TR-024 D.R. Langewisch, G.E. Apostolakis, and D. Helton, Uncertainty and
Sensitivity Analysis for Long-Running Computer Codes: A Critical
Review (January 2010).
MIT-NSP-TR-023 L. Debesse, G.E. Apostolakis, and M.J. Driscoll, The Use of Frequency-
Consequence Curves in Future Reactor Licensing (January 2007).
MIT-NSP-TR-022 N. Haschka and N.E. Todreas, Improving Nuclear Power Plant
Performance: An Assessment of the US Nuclear Fleet Outage
Performance (1990-2005) (July 2006).
MIT-NSP-TR-021 B. Yildiz and M.W. Golay, Development of a Hybrid Intelligent
System for On-line, Real-time Monitoring of Nuclear Power Plant
Operations (April 2006).
MIT-NSP-TR-020 B. D. Middleton and M. W. Golay, Use of Information Theory
Techniques with System Dynamics Models (November 2005).
MIT-NSP-TR-019 A.C. Kadak and T. Matsuo, The Nuclear Industry’s Transition to Risk
Informed Regulation and Operation in the United States; Phase 2
Study for the Tokyo Electric Power Company (August 2005).
MIT-NSP-TR-018 A.C. Kadak, Y. Sakuragi, M.A. Stawicki, and K.F. Hansen,
Quantitative Performance Comparison of Japanese and American
Nuclear Power Industries (January 2004).
MIT-NSP-PR-017 L.W. Hu, J. Lee, P. Saha, and M.S. Kazimi, Thermal Stripping in
LWR Piping System (December 2003).
MIT-NSP-TR-016 G.E. Apostolakis, M.W. Golay, J.P. Koser, A. Knight, G. Sato, and M.E.
Silsdorf, Enhancing the Competitiveness of Nuclear Power Plants
Through Risk Informed Decision Making (November 2002).
MIT-NSP-TR-015 D. Kim and M.W. Golay, Bayesian Belief Network (BBN)-based
Advisory System Development for Steam Generator Replacement
Project Management (November 2002).
MIT-NSP-PR-014 L.-W. Hu, M.S. Kazimi, and A. Sonin, Thermal Striping in LWR
Piping Systems (August 2002).

6. MIT-NSP-TR-013 T. Bannai and M.W. Golay, Probabilistic Risk Analysis of
Falsification Cases in the Japanese Nuclear Industry (August 2002).
MIT-NSP-TR-012 M.E. Silsdorf and M.W. Golay, On the Use of Multi-Attribute Utility
Theory to Quantify the Desirability of Hydraulic Control Unit
Maintenance of a Boiling Water Reactor (August 2002).
MIT-NSP-TR-011 S.M. Oggianu and K.F. Hansen, A System Dynamics Model of the
Energy Policymaking Process (August 2002).
MIT-NSP-TR-010 A.W. Sause and S.M. Oggianu, Modeling Energy Policy Making
Process as a Response to Social Concern (August 2002).
MIT-NSP-TR-009 S.M. Oggianu and K.F. Hansen, Modeling of Electric Energy
Policymaking (March 2002).
MIT-NSP-TR-008 S.M. Oggianu and K.F. Hansen, Modeling the Dynamic Complexity of
the Energy Policymaking Process (March 2002).
MIT-NSP-TR-007 L.-W. Hu, K. Nagasawa, P. Hejzlar, and M.S. Kazimi, Thermal
Striping in LWR Piping Systems (August 2001).
MIT-NSP-TR-006 Y. Sui, M.W. Golay, and K.F. Hansen, Identification of Performance
Indicators for Nuclear Power Plants (September 2001).
MIT-NSP-TR-005 J.J. Kim, M.W. Golay, G.R. Toro, and O. Afolayan-Jejeloye, Analysis of
Direct Seismic Risks and Its Effects on Existing Emergency
Response Plans Near a Nuclear Power Station (October 2001).
MIT-NSP-TR-004 M.R. Galvin and N.E. Todreas, Maintenance Cycle Extension in
Advanced Light Water Design (October 2001).
MIT-NSP-TR-003 B.C. Beer, G.E. Apostolakis, and M.W. Golay, Feasibility
Investigations for Risk-Based Nuclear Safety Regulation (February
2001).
MIT-NSP-TR-002 K. Marchinkowski, R. Weil, and G.E. Apostolakis, CATILaC:
ComputeR Aided Techniques for Identifying Latent Conditions
(April 2001).
MIT-NSP-TR-001 R. Weil and G.E. Apostolakis, Analysis and Utilization of Operating
Experience for Organizational Learning (June 2001).

7.
Premise
The accident at the Fukushima-Daichii nuclear plant has generated worldwide news and
precipitated public concern about the safety of nuclear power in general. The accident has
already caused some governments to re-think their nuclear energy policies, notably including
the Japanese and German governments. There have been calls for cancellation of nuclear
construction projects and reassessments of plant license extensions. This may lead to a global
slow-down of the nuclear enterprise, based on the perception that nuclear energy is not safe
enough. However, the lessons to be drawn from the Fukushima accident are different.
First, the accident was a result of the worst earthquake and tsunami in Japan’s modern history,
an event which has caused the loss of over 20,000 lives and up to $300 billion in damages.
Second, given the extraordinary magnitude of the initiating events (i.e. earthquake was 9.0 vs
design 8.2, tsunami wave was 14 m vs design 5.7 m), the Fukushima-Daichii plant has
performed relatively well in some respects and so far there is no evidence of major human
errors in handling the crisis. It is noted that the containments at Units 1 and 3 have not failed, in
spite of the exceptional loads they have been subject to, i.e. earthquake, tsunami, hydrogen
explosions in the reactor buildings, steam discharges from the reactor pressure vessel,
exposure to hot seawater, pressure above design limits for days. It is likely that there is a leak
in the containment at Unit 2. The release of radioactivity from the plant has been large (with
contributions also from containment venting and spent fuel pool overheating) and some workers
have received significant radiation doses (>100 mSv whole-body equivalent), but health risks for
them and the general population are expected to be negligible (see Appendix A). In fact, no
loss of life has occurred as a result of the accident. Direct damage and casualties inflicted on
Japan by the earthquake and tsunami far exceed any damage caused by the accident at the
nuclear plant. The Fukushima accident has been rated at the maximum level (Level 7) on the
IAEA nuclear event scale, indicating an accident with large release of radioactivity accompanied
by “widespread health and environmental effects”, like Chernobyl. However, there are very
significant differences between Fukushima and Chernobyl. Briefly, the amount of the release
(~10% of Chernobyl), the presence of the containment structures, the radionuclides released
(mostly iodine and cesium isotopes vs. the entire core inventory), the physical form of the
releases (mostly aqueous vs. volatile), the favorable currents and winds at the site, and the
timing of the release with respect to population evacuation resulted in vastly smaller overall
consequences. Having said this, it is important to analyze the technical lessons that can be
learned from Fukushima, so that the safety of nuclear plants in the U.S. and worldwide can be
further enhanced and the attractiveness of nuclear energy sustained over the long term. An
initial attempt to identify the key lessons from the Fukushima accident is presented here.
Objectives
of
the
report
This report presents the reflections of members of the MIT Nuclear Science and Engineering
faculty on the accident at Fukushima, and is offered as a contribution to the debate on the
implications of the accident for the nuclear industry. Our purpose is twofold: we identify and
discuss technical issues arising from the accident; and we begin a review of how the lessons
learned can be used to improve the safety of current and future plants. The information is
organized in six sections: “Emergency Power following Beyond-Design-Basis External Events”,
“Emergency Response to Beyond-Design-Basis External Events”, “Containment”, “Hydrogen
Management”, “Spent Fuel Pools”, “Plant Siting and Site Layout”. For each area, we present
5

8.
key issues observed at Fukushima and corrective actions that should be evaluated for
implementation in current and future plants.
Note
of
Caution
- The technical feasibility and economic impact of the corrective actions discussed in this
report have not yet been fully evaluated; they should therefore not be regarded as
recommendations, but rather as ideas to be explored.
- The situation at the Fukushima plant is still evolving, and not all the information needed for a
detailed reconstruction and analysis of the accident is yet available. The need for and merit
of the corrective actions described in this document should be re-assessed as more
accurate and complete information about the accident becomes available.
- The need for and merit of corrective actions should be evaluated on a plant- and site-
specific basis. For example, it is noted that some U.S. plants already have water-proof
rooms for flooding protection of the diesel generators and related equipment; therefore, the
discussion in Section 1 below would not be very relevant to those plants.
6

9.
1.
Emergency
Power
Following
Beyond-­‐Design-­‐Basis
External
Events
Observations
from
Fukushima:
• The loss of offsite power (due to the earthquake) and onsite AC power (due to the tsunami),
combined with the rapid discharge of the DC batteries led to a complete station blackout,
which in turn led to fuel overheating and damage
Key
question:
How can the station blackout scenario be either prevented or sufficiently mitigated to ensure
minimal consequences?
Possible
corrective
actions
for
current
plants:
• The diesel generators, their fuel, and related switch gear could be housed in rooms at
sufficiently high elevation and/or in water-proof rooms to preserve onsite AC power in case
of tsunamis or floods. Note, however, that seismically-induced stresses increase with
elevation. Interestingly, due to the concern over typhoons and storm surges, all of the
emergency power generation capacity at Korean plants is currently located in water-proof
enclosures, including fuel supplies.
• Utilities and/or FEMA could maintain transportable diesel generators or gas-turbine
generators (i.e. jet engines) that would be rapidly brought to the site (e.g. by air, road or
water) to restore AC power.
Possible
design
improvements
in
future
plants:
• A mix of passive and active safety systems may be desirable to defeat the station blackout
scenario without relying on external intervention. The right mix should be determined
through analysis including risk assessment, taking into account also the possible failure
modes of the passive systems upon occurrence of the initiating external event. A key
question here is: should a mix of passive and active safety systems actually be required in
new plants?
2.
Emergency
Response
to
Beyond-­‐Design-­‐Basis
External
Events
Observations
from
Fukushima:
• There were concerns that TEPCO could not ensure proper staffing of the plant throughout
the accident, if a significant fraction of the local staff had been killed or injured by the
earthquake and tsunami.
7

10.
• The U.S. NRC called for a much larger evacuation zone for U.S. citizens around the
Fukushima plant ("This is the same advice that the NRC would give if this incident were taking
place in the United States, to evacuate beyond a 50-mile radius," NRC Chairman Jaczko, March
17, 2011). While precautionary, this call did not seem consistent with the magnitude of the
radioactivity releases; it undermined the Japanese regulator’s credibility, and created
anxiety and confusion in the media, local population and general public.
• Communication of radiation levels to the public was made difficult by three factors: the use
of three different scientific quantities (dose, dose equivalence and activity), the use of two
systems of units (SI units used worldwide and the older units still in use in the U.S.), and a
lack of context for understanding the meaning of these radiation levels.
Key
questions:
How can proper staffing be assured if a significant fraction of local staff are killed by the initiating
external event? How can the extension of the required evacuation zone be determined when
great direct damage is inflicted on the area surrounding the plant by the initiating external
event? What is the best method to communicate radiation risk to the public in a simple and
effective manner?
Possible
corrective
actions
at
current
and
future
plants:
• A rapid-response team of essential workers could be transported to a stricken plant for
scenarios in which the plant owner/operator cannot staff the plant properly. In the U.S.,
training and operating costs for this rapid-response team could be borne by INPO and/or
consortia of utilities with similar plants, and also assisted by the Air Force for rapid
deployment to the site. The U.S. Federal Aviation Agency (FAA) has a system that may
serve as one model. In countries with a smaller nuclear fleet, the rapid-response team may
even be international.
• Over-conservative evacuation zones (e.g. >20 miles) should not be implemented in case of
accidents initiated by natural catastrophes (e.g. earthquake, tsunami, hurricane) that have
already affected the local population significantly. Large evacuations divert resources away
from the much greater disaster and may create undue stress on the population trying to
cope with the direct consequences of the initiating event. Assessment of the tradeoff
benefits between sheltering and evacuation needs re-emphasis. Evacuation strategies
should be based on minimizing risk to the public from all causes. Extension of evacuation
zones should become a function of both radioactive releases as well as direct damage
inflicted on local area by the initiating event.
• Regulators could demand more on-site personnel to have independent and timely sources
of information, and the ability to influence, or if necessary direct, the owner/operator
behavior during the accident. Note that in such cases where the regulator takes an active
role, the overall responsibility for consequences will then be diffused.
• Radiation risk during nuclear accidents should be communicated to the public using a
qualitative, intuitive scale vs. the traditional quantities of dose rate and activity. For
example, the units of ‘natural background dose equivalence rate’ could be adopted. To
avoid the necessity of adjusting for local background variations, the world average dose-rate
from natural sources should be used: 2.4 mSv/year or 0.27 µSv/hr. Thus the elevated levels
due to contamination would be presented in terms of the factor by which natural background
radiation is exceeded. This approach has several advantages. First, no effort is needed to
8

11.
understand the unit used. For instance, 10 times natural background is easier to grasp than
2.7 µSv/hr since no prior learning in a specialized field is required. Second, there is never a
need to convert between unit systems or to be mindful of numerical prefixes (milli-rem,
micro-Sv, etc.). Third, this method of conveying information about radiation levels reinforces
the concept that some level of radiation exposure is both natural and normal. Finally, use of
this unit implies no estimation of the magnitude of the health hazard from the radiation
levels. This is important since we do not know how hazardous chronic, elevated
background dose rates are, though it is noted that there are regions of the world with
background radiation dose rates one order of magnitude higher than the world-average and
yet with no measureable health consequences.
3.
Hydrogen
Management
Observations
from
Fukushima:
• Deficient fuel cooling resulted in overheating of the fuel, enabling rapid oxidation and
generation of large amounts of hydrogen, which ultimately led to the explosion/destruction of
the reactor buildings at Units 1 and 3, and possibly fires at Unit 4. However, the exact
mechanism of hydrogen accumulation in the reactor buildings has not been ascertained at
this time.
Key
question:
How can hydrogen generation and accumulation be reduced?
Possible
corrective
actions
at
current
and
future
plants:
• Venting of pressure vessels should be via strong pipes connected to the stack (this is
currently a U.S. practice, but it is not clear if it is followed in other countries). Venting should
be possible without power.
• Plants should have the air atmosphere in the pool areas more directly connected to the plant
stacks. Also, fail-open (on power loss) louvers in the buildings could be used.
• More hydrogen recombiners (passive) and igniters (active) could be considered for small
releases in the upper regions of a building, where hydrogen may accumulate. Also, catalytic
recombiners could be used in the ventilation system and inside the containment where it is
not already done now.
• Hydrogen flares for massive venting of containment gases could be explored.
• Use of materials that generate hydrogen upon oxidation with steam could be reduced or
eliminated, e.g., replace Zircaloy cladding with less reactive metals, and ultimately a
ceramic, such as SiC.
9

12.
4.
Containment
Observations
from
Fukushima:
• Due to the station blackout, the operators had to vent (vs cool) the containment to prevent
containment over-pressurization. Some vented gases leaked into the reactor building,
which had no ventilation (again due to the station blackout), resulting in hydrogen
accumulation and ultimately explosion/destruction of the reactor buildings at Units 1 and 3.
Key
question:
How can the need for containment venting be eliminated or its consequences mitigated?
Possible
corrective
actions
at
current
plants:
• The containment should be vented directly to the stack, when containment cooling is not
available. A catalytic recombining system that automatically activates upon loss of power
could also be explored.
Possible
future
improvements:
• Use of passive containment cooling could eliminate the need for venting as a means to
reduce containment pressure, when AC power is not available.
• Use of the filtered/vented containment concept (French-Swedish examples) could provide a
balanced approach to controlling containment pressure and radioactivity releases to the
atmosphere when containment cooling is not available.
5.
Spent
Fuel
Pools
Observations
from
Fukushima:
• Elevated location of the spent fuel pools exposed them to damage from hydrogen
explosions in the reactor buildings at Units 1, 3 and possibly 4.
• Disablement of spent fuel pool cooling may have caused the pool fire at Unit 4 and forced
one-week-long unconventional cooling efforts (e.g. helicopters, water cannons).
Earthquake-induced water leakage from the pools (not confirmed at this time) may have
aggravated the situation.
• The largest radioactivity releases from the Fukushima plant may be from the spent fuel
pools.
Key
questions:
How can the spent fuel pools be better protected from external events? How can reliable
cooling of the spent fuel be ensured in case of station blackout? How can the source term of
the spent fuel pools be reduced?
10

13.
Possible
corrective
actions
at
current
plants:
• Spent fuel assemblies could be moved to dry storage as quickly as possible. Could re-
design dry casks with a “top hat” chimney to enhance air cooling for the hotter fuel
assemblies. However, (i) one must ensure the casks do not tip over due to an earthquake
or hurricane/typhoon, (ii) if the casks are breached, radioactivity release is un-mitigated
(unlike in pools where water provides some scrubbing effect), (iii) the decay heat in pools is
dominated by recently-discharged fuel, so moving the older fuel to dry casks may not have
that significant an impact on pool heat-up time in the event of an accident. These
uncertainties make it unclear whether accelerated dry storage is actually preferable to other
options, such as on-site spent fuel pools or centralized interim storage.
• Current spent fuel pools could be retrofitted with a passive cooling system that can survive
the initiating external event.
• The policy on full core unloading into the pools during refueling shutdowns and spent fuel
pool packing may have to be reviewed.
Possible
future
improvements:
• Spent fuel pools could be housed in containment-like structures separate from the reactor
building. Note that some PWR plants have spent fuel pools inside the actual containment.
• Regional or national consolidated spent fuel interim storage facilities could be built. This
would reduce the spent fuel inventory at the plant, which in turn would reduce the source
term in case of spent fuel pool accidents. Interestingly, Japan has recently completed a
reprocessing plant at Rokkasho and in 10-15 years it is likely that all their spent fuel will be
shipped there rather than stay at reactor sites for long periods of time.
• A national spent fuel repository could be created. The large inventories of spent fuel in U.S.
reactor pools are a consequence of delays of the U.S. repository program that was to have
initiated spent fuel removal from reactor sites by 1998. The U.S. has an operating
geological repository for plutonium wastes generated from defense activities near Carlsbad,
New Mexico, because of a broad national consensus that such a repository was required. A
similar consensus is required for a second repository for spent nuclear fuel.
6.
Plant
Siting
and
Site
Layout
Observations
from
Fukushima:
• Due to this site’s compact layout, problems at one unit created negative safety-related
situations at adjacent units. For example, the hydrogen explosion at Unit 3 disabled some
fire pumps used for seawater injection at Unit 2. Also, it has been suggested that the
fire/explosion at Unit 4 was caused by leakage of hydrogen released from Unit 3 through
shared duct-work with Unit 4. Units 5 and 6, which are far from Units 1-4, were unaffected
by the hydrogen explosions at Units 1 and 3.
• A single external event (the tsunami) disabled all 13 diesel generators at the station
simultaneously.
11

14.
• The Fukushima-Daini and Onagawa plants, both in the vicinity of Fukushima-Daichii,
survived the earthquake and tsunami without major damage.
Key
question:
How can common cause failure and unit-to-unit contagion be prevented?
Possible
corrective
actions
at
current
plants:
• Layout diversity and separation at multi-unit sites could be enhanced. For example, at least
one diesel generator room could be placed sufficiently above grade (for protection against
tsunamis), and one below grade (for protection against plane crashes). Also, in future
plants the administrative buildings and parking lots could be located between units to
enhance physical separation between those units.
Figure
1.
Location
of
current
and
planned
commercial
nuclear
power
plants
(green
dots)
and
all
earthquakes
of
magnitude
≥ 7.0
from
1973
to
2010
(red
dots).
(Figure
courtesy
of
MIT
graduate
student
Mark
Reed)
Possible
future
improvements:
• An obvious approach for future plants would be to choose sites away from highly seismic
areas and coasts, to greatly reduce (and perhaps eliminate) the possibility of damage due to
massive earthquakes, tsunamis and floods. It is noted that people tend to congregate near
coasts and faults (river valleys); therefore, there are strong synergies between minimizing
12

15.
the probability of an adverse external event and maximizing the distance from densely
populated areas. The vast majority of nuclear plants worldwide are already located away
from highly seismic areas (see Figure 1 above). Notable exceptions are the plants in Japan,
Taiwan and California; however, the larger seismic challenge (i.e. higher expected ground
motions) in these regions is currently overcome by a more stringent seismic design of the
plants located in these regions. The strategic question here is: should there be a
requirement to avoid identified vulnerabilities or should plants be allowed to design against
them?
• The number of allowable units at a single plant site could be determined based on an
analysis which accounts for the following, often conflicting, factors: (i) reduction of common
cause vulnerabilities, (ii) availability of staff and resources to address a severe accident
impacting all units simultaneously, (iii) reduction of potential source terms, (iv) high
standardization (shared learning), (v) shared equipment (with implications on both
economics and safety), and (vi) low environmental impact of multi-unit cooling.
A
few
closing
thoughts
The initial response of the nuclear industry and the U.S government to the Fukushima accident
has been measured and rational (see Appendix B). However, the risk of over-reacting to an
accident, particularly one as dramatic as Fukushima, remains high. The industry is concerned
about the near-term effect of Fukushima on the process of life extension of current plants and
the support for new construction projects. Under the pressure of the public and the media, the
government may be compelled to push for sweeping policy and regulatory changes, which may
ultimately prove to be unnecessarily onerous on existing and future plants. Decision-making in
the immediate aftermath of a major crisis is often influenced by emotion. Therefore, the
following questions should be addressed after searching for vulnerabilities at existing plants, but
before enacting significant changes in nuclear energy regulations and policy. Does an accident
like Fukushima, which is so far beyond design basis, really warrant a major overhaul of current
nuclear safety regulations and practices? If so, when is safe safe enough? Where do we draw
the line? It seems that a rational approach to this question would need to be based on a risk-
informed comparison of nuclear energy with other energy sources (particularly its most credible
competitors, such as coal and natural gas), including their effects on climate change, global
economy, stability and reliability of the energy supply, and geo-politics. But can the decision
makers take a risk-informed approach to energy policy?
All engineered structures (e.g. power plants, bridges, skyscrapers, dams, highways) will fail if
subjected to loads far enough beyond what they were designed for. The catastrophic failure of
an irrigation water dam in the Fukushima prefecture, which occurred when the earthquake hit,
went virtually un-reported in the media. What does this failure say about the safety of hydro
power? Are the design basis selections of energy industry structures posing high environmental
hazard, such as oil drilling platforms offshore, coal mines and water dams, consistent with those
of nuclear plants? If not, are we as a society irrationally accepting higher risks from certain
technologies than others?
13

16.
2

17.
APPENDIX
A
–
PUBLIC
HEALTH
IMPACT
OF
FUKUSHIMA
Radionuclides
of
Concern
While there are many radionuclides that can be released at the time of a reactor accident, not all
have the potential to impact public health because of issues related to: abundance, decay
scheme, half-life, and chemistry (which ultimately affects route into the body, anatomical area of
concentration, and residence time). Noble gases such as krypton and xenon rapidly disperse in
the atmosphere; heavy elements are non-volatile so, if released outside the containment, tend
to stay at the plant or in the near vicinity. The isotopes of particular concern are 131I and 137Cs.
Both decay by a combination of beta and gamma emission, which means they can represent
both an internal and an external hazard. They are released in relatively high abundance and
their half-lives (8 days and 30 years, respectively) are sufficiently long that they do not decay
before being widely distributed in the local environment, yet are sufficiently short that enough
nuclei will decay to result in significant and measureable doses in the time scales important to
human life.
Measured external gamma dose-rates following the tsunami and subsequent damage to the
cooling systems at the Daiichi nuclear power plants spiked on March 15 and 16 and thereafter
gradually declined. The rate of decline is a result of the combined effects of environmental
dispersion and physical decay with a mix of the short half-life 131I and the much longer half-life
137
Cs. Nine weeks after the emission spike the effective half-life of the measured gamma dose-
rate is approximately 70 days. The effective half-life continues to increase but will always be
smaller than the physical half-life of 137Cs due to effects of weathering and further distribution in
the environment. Peak gamma dose rates at different geographical locations depended on both
distance from the damaged plant and on wind and rain patterns. Iodine and Cs reach the
ground via dry deposition but deposition is hastened by rainfall which can lead to local areas of
high activity. Wet and dry deposition onto crops and subsequent human ingestion, or ingestion
by cattle followed by consumption of contaminated milk, is the most common route into the
body. Radioiodine was of most concern in the immediate aftermath of the accident both from an
external dose perspective and because of the potential for induction of thyroid cancer,
particularly in children (internal dose). Drinking water restrictions based on 131I levels were in
place for a number of days, particularly for infants for whom a maximum level of 100 Bq/L was
recommended1.
It is 137Cs that represents the most significant long-term hazard of a contaminated environment.
Chemically it behaves like potassium which is found in all of our cells, so it is readily taken up
and used if available. Like iodine it will settle out of the radioactive cloud onto fields and crops.
Since it binds tightly to moist soil it is not readily taken up via the root structures of plants
however it can enter plants upon falling onto the surface of leaves. Elevated levels of 137Cs in
1
In the U.S., risks from inhaling or ingesting radioactive isotopes are tabulated in the EPA's "Cancer Risk Coefficients
for Environmental Exposure to Radionuclides, Federal Guidance Report No. 13 (EPA 402-R-99-001)" published in
131 -10
1999. The risk coefficient for fatal cancer from I is 1.85 x 10 per Bq. Using this number and assuming lifelong
-10
consumption of 1.2 kg of food contaminated at a level of 100 Bq/kg we calculate a risk of: 1.85 x 10 /Bq x 1.2 kg/d x
4
2.75 x 10 d x 100 Bq/kg = 0.00061. This risk is tiny compared to the ‘natural’ risk of cancer and very small compared
131
to other risks encountered on a daily basis. It is also a very conservative estimate of the risk since I has
substantially decayed and lifetime consumption of food at this contamination level will not occur. For these reasons
an activity level of 100 Bq/L is considered safe for infants.
14

18.
several foodstuffs required restrictions on consumption and prompted a number of countries to
limit imports from Japan for some time. All drinking water interdictions were lifted in early May
however several foodstuffs still show radiation levels that exceed regulation values set by
Japanese authorities.
Radiation
Doses
Attempts are ongoing to keep the cumulative radiation doses to the Japanese public below 20
mSv in the first year following the reactor accident. [Doses will be substantially lower in
subsequent years due to environmental dispersion and physical decay of residual 137Cs.] This
effort involves (i) monitoring radioactivity levels in foodstuffs and water and prohibiting sale and
consumption where necessary , (ii) recommending sheltering indoors in areas where cumulative
dose-rates over one year are expected to be > 10 mSv, and (iii) relocation of residents from
within a 20 km radius zone around the plant. 70,000-80,000 residents were relocated in the first
month after the accident but relocations are continuing in areas where residents are predicted to
receive doses in excess of 20 mSv in the first 12 month period.
Doses to people living further from the Daiichi plant are much lower. In Tokyo, 240 miles away,
residents can expect an additional cumulative radiation dose of 1 mSv from the first year, a 40
% addition to the 2.4 mSv they already receive from natural sources. As of the first week of
May, external gamma-dose rates in Tokyo are 0.09 µSv/hr, a factor of almost two above natural
levels (0.05 µSv/hr). Since external gamma dose contributes ~ 20% to the total background
dose (the remaining dose components are cosmic rays, internal radionuclides, and radon
daughters), this increase in gamma ray exposure currently adds 16% to the daily radiation dose
to Tokyo residents.
Health
Implications
The impact of low doses of radiation on our health is assumed to be an increase in the
probability of being diagnosed with cancer. No other natural disease shows a significant
elevation following exposure to low dose radiation and no unusual or unique diseases are
created. Radiation-induced cancers have a latent-period of 20-30 years (shorter for leukemia)
and tend to appear at the same time in irradiated as in unirradiated populations. Since the
cancers induced by radiation are the same types of cancers observed ‘naturally’, determining
the number of additional cancers caused by a small dose of radiation when baseline cancer
rates are already high has not been possible for doses in the 20 mSv range (or even higher).
Although no data have ever demonstrated that 20 mSv over 1 year results in measureable
harm, this dose range has long been relevant to the occupational radiation protection field and
thus there has been a need to generate radiation risk estimates, even in the absence of actual
data. These estimates come primarily from the long-term evaluation of the A-bomb survivor
population and are a result of adopting a hypothetical model of extrapolating the risk per unit
dose at high dose levels down to the low dose range. Use of this extrapolation model in the
generating of risk estimates incorporates a number of assumptions appropriate to radiation
protection in the workplace but not appropriate to determining the hazards of an environment
contaminated with a long-lived radionuclide. Accordingly, scientific bodies evaluating risk often
specifically caution against extending these strategies to predicting the long term effects of
small doses to a large population. Unfortunately, more applicable risk estimates do not exist
and so this caution is routinely ignored when the potential impact of low doses is of interest.
15

19.
The U.S. National Academies of Sciences’ BEIR VII committee estimates that 1 cancer could
result if 100 people received a single dose of 0.1 Sv (a risk of 0.01/0.1 Sv), with lower doses
resulting in proportionally lower risk. Thus, a dose of 20 mSv (if delivered acutely) x 0.1 per 0.1
Sv = 0.002. In other words, the 20 mSv dose ceiling pursued by the Japanese authorities
represents a 0.2 % chance of being diagnosed with cancer later in life, in addition to a 42 % risk
an individual already faces from ‘natural’ causes. This estimate is expected to be high by a
factor of 2-10 and possibly more, according to NCRP 64, to account for the reduced impact of
protracted radiation delivery, relative to the same dose received all at once.
20 mSv over the course of a year represents a factor of 8 times the average natural radiation
background level. It is the equivalent to 2-3 abdominal CT exams for a lean individual, or
equivalent to one CT exam for someone who is overweight. However 20 mSv received over the
course of one year is expected to have less biological impact than the same dose received via
medical imaging since the dose is protracted over time.
The
Cost
of
Dose
Avoidance
Permanent and long-term relocation can reduce exposure to radiation to essentially zero levels
above natural background. What is gained is the elimination of a tiny additional risk of cancer
(maximum risk of 42.2 % instead of 42.0 % at 20 mSv). This cancer, if it appears, will be
diagnosed many years, perhaps decades, in the future. But this gain comes with very
significant costs. The costs include loss of home or farm (48,000 homes and over 400 livestock
or dairy-farming households are in the evacuation region), loss of privacy (shelters are crowded
and residence time is expected to be measured in months before alternative temporary housing
will be available), and loss of community (whole towns and villages have been evacuated).
Prohibition against consuming contaminated food and water results in no additional internal
dose but, for a country already facing food shortages following a devastating earthquake and
tsunami, the loss of valuable foodstuffs and interdiction of farmlands are a significant price to
pay.
The costs of dose avoidance are high. A clearer understanding of the actual risks represented
by, say 20 mSv, would help residents and government officials engage in a productive dialogue
regarding how to make the tradeoff between dose avoidance and loss of important aspects of
daily life (home, food, and community). It is also critical that the public gain a wider
understanding of the bases on which our radiation risk estimates are derived. The inherent
protection of radiation workers built into our estimates of radiation risk have been effective in
ensuring that employers keep dose to their workers very low, and thereby the need to actually
know the hazard from radiation levels that are 5, 10, or even 50 times background has been
avoided. However, this approach is not useful in the situation of a contaminated background
where conservative estimates of risk force residents to make significant sacrifices to avoid all
dose.
It is also important that residents understand the manner in which protection limits are based on
risk estimates. For instance, a limit imposed on employers to restrict exposure of the general
public does not correspond to a declaration that doses below this limit are safe but above this
limit are not. Concerns have been raised regarding elevated dose-rates at schools in
Fukushima prefecture, almost 170 of which have been forced to relocate or close. Raising the
maximum allowed annual radiation limit from 1 mSv to 20 mSv in schools led to a significant
uproar and prompted one government advisor to resign in protest. Governmental ministers
defended the increase from 1 mSv to 20 mSv/year as a necessary measure to guarantee the
education of hundreds of thousands of children in Fukushima prefecture. However many
16

20.
members of the public viewed this step as regulators changing their mind regarding what levels
are safe, rather than seeing the situation as a choice between two undesirable situations. Given
that the environment has been contaminated, the choice to residents of Fukushima prefecture
involves accepting the 0.2% additional chance of getting cancer in 15-30 years, or delaying the
resumption of normal schooling (and a normal life), for an extended period of time.
In the United States the EPA recommends implementation of a return home dose rate that
would lead to a maximum dose of 20 mSv in the first year following a reactor accident; many
states have adopted this recommendation. This is the same level that has prompted such
emotional response from frightened members of the public and even from advisors to the
government during the on-going crisis in Japan. Once an accident has taken place and the
environment is contaminated, we need to be equipped with the most accurate estimates
possible of harm from living with elevated background radiation levels. These can then be
weighed against the benefits and drawbacks of dose avoidance strategies. We are not there
yet.
17

21.
APPENDIX
B
-­‐
RESPONSE
OF
THE
U.S.
NRC
AND
NUCLEAR
INDUSTRY
TO
THE
FUKUSHIMA
ACCIDENT
The U.S. NRC and nuclear industry are responding to the accident in Japan, focusing mostly on
the issues discussed in this document. In particular:
- The NRC has established a task force to do a 90-day review of lessons learned from
Fukushima, including prominently the issues of protection against natural disasters,
response to station blackouts, severe accidents and spent fuel accident progression. On
the issue of station blackout, it is noted that, per the NRC rule of 1988, U.S. plants must
determine the coping time, i.e. the time the plant can survive without AC power. Typical
coping times at U.S. plants are currently between 4 and 12 hours, determined by the battery
capacity. This establishes the time window within which either AC power must be restored
or new batteries brought to the site. The adequacy of the current coping times is being
reviewed.
- U.S. nuclear utilities have approved an industry-wide assessment, to be completed within 30
days under the leadership of INPO, to verify and validate each plant site’s readiness to
manage extreme events. The assessment includes the following actions:
• Verifying each plant’s capability to manage major challenges, such as aircraft impacts
and losses of large areas of the plant due to natural events, fires or explosions. Specific
actions include testing and inspecting equipment required to mitigate these events, and
verifying that qualifications of operators and support staff required to implement them are
current.
• Verifying each plants capability to manage a total loss of off-site power. This will require
verification that all required materials are adequate and properly staged and that
procedures are in place, and focusing operator training on these extreme events.
• Verifying the capability to mitigate flooding and the impact of floods on systems inside
and outside the plant. Specific actions include verifying required materials and
equipment are properly located to protect them from flood.
• Performing walk-downs and inspection of important equipment needed to respond
successfully to extreme events like fires and flood. This work will include analysis to
identify any potential that equipment functions could be lost during seismic events
appropriate for the site, and development of strategies to mitigate any potential
vulnerabilities.
- TVA owns the Browns Ferry plant with three BWR units featuring Mark-1 type containments,
similar to the Fukushima-Daichii plant design. TVA has stated that the plants already have
explosion-resistant pipes to vent hydrogen from the containment, fire-hoses pre-placed to fill
spent fuel pools in case of loss of cooling, and hardened diesel rooms, including 7-day
supply of fuel, behind water-tight doors. The diesel switchgear is located within the reactor
building, and thus is protected from flooding. Also, as a result of Fukushima, TVA has
bought diesel-driven fire pumps. Most U.S. plants also have battery carts located
throughout the reactor building to provide power for critical valves and instruments, should
AC power be lost for a few hours (i.e. their coping time, as defined above).
Finally, we note that operators of nuclear power stations in Japan have been urged by the
Japanese government to ensure their facilities have back-up emergency power sources. The
18

22.
government told utility companies they should have mobile generators on hand to cool their
nuclear reactors as an added safety measure. The utilities have been asked to confirm the
steps they have taken and conduct drills within a month or stop operating their nuclear facilities.
19